Title: UCLlogo.EP Creator: Adobe Illustrator(TM) 1.1 Preview: This EPS picture was not saved with a preview included in it. Comment: This EPS picture will print to a PostScript printer, but not to other types of printers. Title: Constitution logo paths Creator: FreeHand 5.5 Regional Government and Public Health by Scott Greer and Mark Sandford This briefing was funded by: The Leverhulme Trust It was launched at a seminar hosted by: November 2001 Published by The Constitution Unit School of Public Policy UCL (University College London) 29/30 Tavistock Square London WC1H 9QU Tel: 020 7679 4977 Fax: 020 7679 4978 Email: constitution@ucl.ac.uk Web: www.ucl.ac.uk/constitution-unit/ © Constitution Unit, UCL 2001 This report is sold subject to the condition that is shall not, by way of trade or otherwise, be lent, hired out or otherwise circulated without the publisher’s prior consent in any form of binding or cover other than that in which it is published and without a similar condition including this condition being imposed on the subsequent purchaser. 2 Published November 2001 3 Contents EXECUTIVE SUMMARY..........................................................................................................................5 INTRODUCTION ......................................................................................................................................6 BACKGROUND ........................................................................................................................................7 MAKING HEALTH WORK ........................................................................................................................8 London ............................................................................................................................................10 GLA Structures ...............................................................................................................................11 London governance bodies outside the GLA .................................................................................13 The Government Office for London. .......................................................................................13 Whitehall and the Department of Health. ................................................................................13 The NHS London Region. .......................................................................................................13 The London Health Commission ....................................................................................................14 Summary: the actors.......................................................................................................................15 EAST MIDLANDS...................................................................................................................................16 Regional structures .........................................................................................................................17 emda .......................................................................................................................................17 The Regional Chamber ...........................................................................................................18 The NHS .................................................................................................................................19 HOW IT WORKS ....................................................................................................................................19 Soft money ..............................................................................................................................20 Secondments ..........................................................................................................................21 Networking ..............................................................................................................................21 Shields.....................................................................................................................................22 WHY DEVOLVE PUBLIC HEALTH?......................................................................................................23 1. Regional public health would build on existing success .............................................................23 2. Regional public health would fit with a broader role for regions .................................................24 3. Regional public health would integrate health with other policies ..............................................25 4. Regional public health would fit with the developing structure of public health in the NHS .......26 BLUEPRINTS .........................................................................................................................................26 The public health competency ........................................................................................................27 Transfer of staff and budgets..........................................................................................................27 Regions should coordinate with the centre and have a veto over central public health initiatives.28 Ideas for the regions .......................................................................................................................29 UNELECTED CHAMBERS ....................................................................................................................30 CONCLUSION........................................................................................................................................32 BIBLIOGRAPHY.....................................................................................................................................32 4 Executive Summary English regionalism will not progress in the same way as devolution to Scotland, Wales and Northern Ireland. Neither the institutions nor the political will are well enough developed. This is particularly true of the National Health Service, a front-line public service, on which the Labour Government has staked its reputation. Regional government has discovered public health. A new tier of governance searching for functions has found a Cinderella function searching for a new home. This thesis is explored in two regions: London and the East Midlands. London is the less successful of the two regions in joint working for public health. An exception is the London Health Commission, an advocacy coalition of some 40 chief executives meeting regularly to discuss public health. It has built health impact assessments (HIAs) into the Mayor’s strategies on subjects as diverse as transport and culture. But the influence on London governance has been less notable. The East Midlands has little previous history of joint working, and is distant from the preoccupations of national media. As a result the regional bodies have greater latitude for innovation. They are more permeable to policy ideas from outside, typically delivered by secondees. Regional bodies and soft money can permit the use of resources for aims other than meeting centrally-set performance indicators, and create a space in which staff can do so without fear of central disapproval. Public health functions and networks, built up by the voluntary arrangements in East Midlands and London, could be transferred to elected assemblies in all English regions. The staff and budgetary requirements are small, but the potential rewards considerably larger. Existing interests in the NHS, and at central and local level, are not threatened. Public health policy works best through a consensual, network-driven approach. Joined-up government of this kind is one of the greatest advantages of democratising the English regional tier. Staff and budgets could transfer from the newly-created Government Office teams, the Health Development Agency, and the Public Health Observatories. Better still would be the inclusion of Single Regeneration Budget and New Deal for Communities funds. Whitehall control, in the form of indicators and targets, should be as loose as possible. Those regions without directly elected assemblies in the near future should encourage secondments, prepare public health plans, and give public health greater salience in existing work. 5 Introduction Why devolve anything at all? Before advocating the devolution of new powers, it pays to recall the reasons for devolution. Regardless of where they are, devolved governments offer the integration of public services, the development of innovative regional strategies, the democratisation of policy and the adaptation of policy to place. Devolution to the English regions offers all of these advantages. The question for policymakers is how to design an English regional settlement that gains these advantages. We argue that one policy field in which regions can demonstrate their advantages is that of public health, and that public health competencies should be devolved to elected regional assemblies in England. English regional devolution cannot be a process like we saw in Northern Ireland, Scotland or Wales. Devolution until now has been about creating legislative bodies in pre-existing territorial administrations—the Scottish Parliament and the Welsh and Northern Irish assemblies each inherited most of their competencies from the pre-existing Scottish, Welsh, and Northern Ireland Offices and the Northern Ireland Civil Service. The reach and resources of the administrations barely changed; it was the structure of accountability that changed. The borders of the devolution settlement were largely settled before the arrival of the elected bodies. By contrast, devolution within England has no previously fixed map of regional powers. There is no clear layer of regional administration, akin to the Scottish Office or Northern Ireland Civil Service, that could have a democratically-elected assembly attached. Thus the process of political devolution must go with the process of constructing the region as a political unit. In other words, devolution to English regions must come with the definition of English regions as real political units. Given that England is a highly centralised political unit, designing regional governments will require careful thought it the regions are to neither damage the workings of public administration nor be pointless. We propose that public health be a function of elected regional assemblies. Turning over extensive public health powers to the regions would: build on the existing successful regional working in public health, working which has already demonstrated the value of a regional public health focus fit with the growing role of regions in coordinating, integrating, and adapting economic and social policies fill in a gap in British social provision by integrating health with other policies fit with the developing structure of public health in the NHS. The rest of this study documents the extent of current regional working and its virtues; the way public health working interacts with its institutional environment; and the mechanisms necessary to integrate public health with ongoing regional activities in social and economic development and inclusion. 6 Background This paper brings together two longstanding interests of the Constitution Unit, which have both been very active in 2001. The first is the Unit's work on regional government in England, now led by Mark Sandford. This has resulted in two major reports this year, the first on elected regional assemblies, Unexplored Territory: Regional Assemblies in England (July 2001); and the second on regional chambers, Further Steps for Regional Chambers (December 2001). The second longstanding field of interest is our work on Devolution and Health, funded by the Nuffield Trust. In May 2001 Scott Greer joined us to work full time on additional projects on devolution and health, funded by the Leverhulme Trust. He soon found that his interest in the devolution of health policy in England overlapped with Mark Sandford's interests in regional government. This research paper is the result. It has been funded by The Leverhulme Trust, and was launched at a seminar on 12 December 2001, hosted by The Nuffield Trust. The present paper grew out of attempts to study the effects of the nascent structures of governance in the English regions on health policy. Unlike in Northern Ireland, Scotland and Wales, health services proper within England will not be devolved. However, the regional organisations of England1, including London, have begun to develop integrated policies in the previously ignored field of public health. These policies were the subject of a recent study by Scott Greer, The Real Regional Health Agenda, which investigated developments in the East Midlands. Health, as one of—perhaps the foremost of—the public services, is inevitably drawn into the developing debate about the introduction of elected regional assemblies to England. The Government’s white paper on English regional government is due out in early 2002: there has been some, albeit limited, speculation that it may propose a model of regionalism comparable to the Greater London Authority—a body charged with writing strategies and conducting scrutiny. Thus, we decided to compare the conclusions of Scott Greer’s study of the East Midlands with the situation in London, which possesses its own unique, and curious, elected government. The comparative study focuses on two issues: what builds successful regional working; and what set of institutions most effectively promotes regional success and accountability? Based on our conclusions, this report makes suggestions, in the light of the investigation into current practice, of how and why public health would be an ideal responsibility for new elected assemblies in the English regions to take on. We believe this is a ground-breaking exercise in delineating in detail what ‘devolution to the English regions’ would mean for current institutions and policy programmes, and we hope that the level of detail will reduce 1 For an introduction to these, see Mark Sandford, Further Steps for Regional Chambers, Constitution Unit, 2001 7 the ‘fear of the unknown’ which must inevitably dog the development of the regional White Paper. Making health work Despite the lack of serious regionalism in England, regional public health joint working has emerged—something that testifies both to its intrinsic virtues as a channel for social and economic policy, and to the commitment of activists from across different institutions who see ways to incorporate public health outcomes into their own agendas, to general benefit. From the mid-1990s onward, public health professionals, policymakers, and decision makers in other areas began to formulate and push an agenda in which public health and the wider determinants of health would be serious issues for policy outside the NHS. The intellectual basis for these theses is simple and nearly irrefutable. Most great improvements in morbidity and mortality in the modern world have been attributable to work outside acute care. Sewers and clean drinking water, public transport and air quality regulation, immunisations and screening, jobs and training, education and healthy habits all have effects on quality and length of life that are rivalled only by a few of history’s medical breakthroughs. Thus, the best and cheapest way to solve many of England’s serious health problems is not through more investment in expensive acute services that will take years to come on line; it is through preventive programmes that reduce the need for such services now and in the future. This agenda faces three obstacles. First, there is a strong cultural sentiment that health is “what the NHS does.” On one side, those inside the Service face a medical professional culture that exalts the academic and hospital specialists who work with some of the toughest diagnostic and treatment issues rather than those who grapple with disease vector analysis or dietary education. On the other side, much of local government and the state remains oblivious to the health consequences of their strategies—not realising that decisions about bus services, business parks, school curricula or domestic abuse plans have major health consequences and might not achieve their goals without major health inputs. Second, in many cases there is overlaid on this divide an atmosphere of deep suspicion and dislike in relations between the NHS and other parts of local and national government. The NHS began by taking over hospitals once owned by local governments, and as a centralised, professional organisation has long had problems dealing with democratic local government. Furthermore, the core of the NHS is in health services, not public health. Until recently, there was little institutional support for attention to health beyond the health service. Meanwhile, every other policy field already has bureaucracies at work, with their own ways of planning policy. Thus, bringing health into their concerns requires that they change in order to achieve a goal they used to think was not their responsibility. 8 Third, every government agency remains subject to demands for delivery, which often means transferring resources and energy from policies with major long-run consequences to policies with smaller but more visible short-run consequences. Thus, waiting lists for the NHS or demands on others for measurable outcomes push resources away from the long-term, cumulative work needed to solve structural problems. Public health is particularly vulnerable to sacrifice, because many of its effects are very long-term: reducing teen pregnancy in a year will only show its effects on the health service in decades, when there are fewer chronically ill low-birth-weight babies born to very young mothers. Thus, public servants who care about public health and see its advantages must fight to preserve the necessary resources from short-term demands for more delivery. To surmount these problems, the new public health agenda in the regions has developed a particular profile that allows it to move forward. First, it has an organisational demi-urge in the public health specialists of the NHS regions, the Health Development Agency, and many Health Authorities. This group carries the baton, maintaining interest and constantly seeking ways to build policy networks and policies. Second, to solve the problems of disconnection between the NHS and other agencies, these activists work to build networks and establish forums and groups that allow those dedicated to health to educate each other and develop collaboration to produce health. Given the centre’s relative tolerance of and even support for joint working and public health, they have been able to scrape together resources to engage in these activities, and perhaps win allies in the fight for resources for their long-term goals. In the two regions we studied, this strategy was a good fit with regionalism. In both cases, the regional government (the GLA and the East Midlands regional bodies) was interested in developing new policy ideas and demonstrating the benefits of a regional approach to potential stakeholders and allies. Public health fitted the bill, and the public health specialists across the NHS were there to sell it. They could propose a way to bring together different regional actors, a way to integrate disparate strategies, and a way to create new policy networks and communities that could become the basis for regional politics. On the regional level, they could offer real activity in the form of policy integration, with health specialists assessing and commenting on plans to see how health benefits might be squeezed out of a transport, educational, or planning project. On the ground, they were establishing concrete projects of all sorts through collaboration with local government and others, and they could point to this as the sort of joint working that regional collaboration could intensify and improve. A marriage was possible between a policy looking for an institutional home and a set of institutions looking for policies. The rest of this section discusses the course of events and the outcomes to date in two different social, political, and economic environments. In London, the crucial event marking health policy has been the creation of the elected Greater London Assembly and the Mayor of London, with powers over the Greater London Authority. The GLA was an important environmental disturbance, and all organisations immediately began to examine it for 9 potential dangers and opportunities. A legacy of fragmentation and suspicion, exacerbated by the polemics in the mayoral election, has slowed the creation of policy communities around it. It is nevertheless demonstrating the value of regional-level policy coordination in the educational and policy ventures that have begun under its wing. By contrast, the East Midlands is a typical English region, with weak statutory institutions and a short history of collaboration. However, its public health and regional activists have built a strong policy community with the resources they have, improving the profile and performance of the regional institutions while bringing policies together. In each case, the regional arena has proved valuable; the differences between the two regions are instructive for policy design. London London is frequently the forgotten ninth region of England. Commentary and analysis on ‘the English regions’ often refer to the eight regions outside of London, without acknowledgement of the existence of a ninth with a distinctive, and distinctly odd, polity. London’s status as both a large urban area, a region in population terms and for the purposes of many government regional offices, as the location of the seat of the UK Government and a vast number of business central offices, do of course mean that London is a region like no other in the UK, and this helps to explain not all, but many, of the features which distinguish its development from the East Midlands (or any other region of England). The influence of history on the development of London’s governance is vital to understanding recent institutional events. London was governed, from 1964 to 1986, by the Greater London Council, covering the same area as today’s Greater London Authority. This was an upper-tier local authority with some unusual features, and was designated a ‘metropolitan council’ in the 1974 local government reforms. Large swings in the political balance of the council were the norm at election times. The GLC was complemented by 32 London boroughs, plus the Corporation of London, which represented the lower tier of local government. But the GLC “was heavily dependent upon the boroughs for successful implementation of whatever programmes it chose to pursue, and was constrained by central government”.2 It was always vulnerable to the charge of being redundant, having made little impact until the election in 1981 of a young, radical and very visible Labour leader in Ken Livingstone. Livingstone’s policies subsequently attracted the ‘loony left’ tag, and the ire of the then Conservative government, providing a political excuse for the GLC’s 1986 abolition. Since 1986 the boroughs have been unitary local authorities with all local government functions competencies. 2 Brendan O’Leary, “London Governance: Past,. Present and Future”, in eds Michael Hebbert and Tony Travers, The London Government Handbook, Cassell, London, 1988 10 The Labour Party came to power in 1997 with a clear commitment to reintroducing London-wide government. But it was anxious to avoid comparisons with the embarrassing past of Livingstone’s GLC: especially as it had become clear by 1999 that Livingstone, by then an MP, would want to run for the new position of Mayor of London. Thus, the Greater London Authority was explicitly a very different body, bringing together a number of experiments in governmental practice. GLA Structures The GLA has Britain’s first directly-elected mayor with sole executive power. The Mayor is elected by the Supplementary Vote. He or she may appoint an advisory cabinet, but takes all the GLA executive decisions (with one exception below). The majority (90%, or £3.3bn) of the Mayor’s budget passes to the four ‘functional bodies’: the Metropolitan Police Authority, the London Development Agency, Transport for London and the London Fire and Emergency Planning Authority. These bodies’ budgets are allocated by the Government. The Mayor is then able to raise a precept on the London boroughs, in order to fund GLA services. This precept amalgamates within it previous precepts levied by the Metropolitan Police and the Fire and Emergency Planning authority. The remainder of the precept funds GLA running costs and various very small GLA programmes. A breakdown of the GLA budget is shown in Table 1. The Mayor has an arm’s-length relationship with the functional bodies. He appoints some of the board members of each: for the MPA and LFEPA, a majority must be elected GLA members or borough councillors. Day-to-day management rests with the executives of the bodies. Thus, Table 1 demonstrates the tight limits on the Mayor’s freedoms: the defining characteristic of this model of regional government is the absence of executive power. The Mayor is complemented by a directly-elected Greater London Assembly of 25 members. 14 are elected by the First Past the Post system, in constituencies of two or three boroughs; the other 11 are elected on an Additional Member top-up list covering the whole of London. The Assembly’s role is to scrutinise the Mayor and the functional bodies, and it can reject the Mayor’s budget by a two-thirds majority. 11 Table 1: The Greater London Authority budget Core GLA expenditure: consisting of- £36,082,000 Mayor’s Office £1,623,000 Assembly and Secretariat + LTUC £6,741,000 Election provision £3,500,000 Chief Exec £717,000 Communications £2,910,000 Finance & Performance £3,880,000 Policy & Partnerships (inc. Squares) £10,935,000 Corporate Services £7,094,000 Overheads £3,058,000 Income (inc. savings & interest) (£4,376,000) £36,082,000 Functional bodies London Development Agency £308,373,000 Transport for London £781,200,000 Metropolitan Police £2,258,312,000 LFEPA £319,000,000 Total budget £3,702,967,000 % which Mayor can vire 0.97% The Mayor has no direct powers over health services. However, he or she does have a duty, according to the GLA act, “to promote improvements in the health of persons in Greater London”.3 Originally, the act had no health responsibilities for the mayor, as the NHS and others were fearful that the mayor would try to take on a role in direct governance of health services. As the discussions leading to the act progressed, however, public health specialists at a number of key London institutions were able to persuade the NHS that the Mayor could help with public health (that once again, public health could provide a common ground between political forces that mistrust each other) and to persuade the London politicians that a public health competency was a good thing to have. Ken Livingstone made a ‘health strategy’ part of his election campaign: the GLA has six health staff and a budget of £200,000 (excluding salaries). Signally, one of his appointed advisors is the director of the NHS London Region public health function; several interviewees noted that public health was one of the few policy areas in which the distrust between the government and the GLA could be overcome for fruitful work. 3 Greater London Authority Act 1999, s30 (5) (a) 12 London governance bodies outside the GLA As stated, the salient attribute of the GLA is the absence of executive power. The Mayor must govern through influence and negotiation. Most executive power in London continues to lie in similar places to where it did before the GLA’s arrival. For the purposes of public health, the important players are: The Government Office for London. This is one of the nine Government Offices, initially set up in 1994. For most respondents it has had a shadowy existence. Briefly before the setup of the GLA, one individual within GOL took a lead on public health, and built up some effective relationships. This individual retired, and nobody took his place. Health is not a Government Office responsibility, so it was very vulnerable to the personalisation of policy initiation and the current instability of GOL functions. Whitehall and the Department of Health. Respondents did not feel that the DoH was, or should be expected to be, a major player in London’s health. The most useful function it could perform was to loosen the very tight management of the NHS, via performance targets of all kinds. There was criticism from one respondent that the DoH’s policy towards London was irrationally coloured by ‘fear of Ken’ [Livingstone]. The NHS London Region. The NHS has been an important driver of the whole London health agenda. In particular the work of the London regional Director of Public Health has been important. London is also fortunate in that it is the only regional office unaffected by the proposals in Shifting the Balance of Power,4 meaning that NHS functions currently organised around the city will remain so while they are shuffled around in the rest of the country. There should be less disruption than elsewhere, and less damage to the networks that make integrated policy working happen. There are others. As in the rest of the English regions, the GLA has begun to build relationships and partnerships with the multiplicity of quangos, voluntary bodies, and businesses. But the nature of the GLA lends this process a different emphasis in London. The East Midlands structures could be described as ‘organic’: they grew out of previously existing networks and organisations. The GLA, on the other hand, has been grafted on to the governance of London. It has neither significantly reduced fragmentation, nor does it fit neatly into an existing gap. The Greater London Assembly in particular has struggled to find a role. Most health actors are unsure how to relate to the GLA; what it can do for them; how it works as an organisation; or how influential politics is in its working, and the GLA itself 4 Department of Health, Shifting the Balance within the NHS: Securing Delivery, London, 2001; this is the White Paper detailing the latest reorganisation of the NHS. 13 has done little to reduce this ignorance. One organisation, however, deserves pride of place in the structure of London health policy: the London Health Commission. In a fragmented policy arena riven by distrust and uncertainty about powers and alliances, it provides a space for education and policy coordination and is creating a London health policy community. The London Health Commission In a polity which functions through influence and negotiation, a natural means of working is through the advocacy coalition.5 This is an informal group of as wide a range of experts in the field as possible, which meets to discuss and advocate policies (the concept constantly reappears under different names in political science, as researchers encounter it across countries, policies, and epochs). What has happened in London public health (and to a lesser extent in the other regions) is that the advocacy coalition has been semi-formalised. This semi-formal advocacy coalition is the London Health Commission. The London Health Commission grew initially from the Turnberg Review of 1999, which created, for the first time, a London NHS region on the Greater London boundaries. Several organisations—the King’s Fund, Government Office for London, Association of London Government, Metropolitan Police, the NHS, the London Development Agency, Health Development Agency and others—provided sponsorship for the Commission (it appears that the NHS London region leadership and the King’s Fund were the organisations that provided the initial bases for the idea champions). The Commission was also able to influence the GLA’s health policy, though not that of London as a whole, by dint of having been set up earlier than the GLA. In that sense it was able to occupy the ‘empty chair’ of Mayoral health policy6. The Commission had a health strategy in place by the time of the Mayoral election in May 2000. As a result, the London health agenda has borne down on the points stressed by the Commission, namely the need to consider health impacts of other policies, and the need to develop links between the NHS and other policy fields. The Commission’s most important contribution has been to carry out Health Impact Assessments on various of the Mayor’s strategies; the highest-profile strategy so far audited is the Transport Strategy. The strategies produced by the Mayor are a significant part of his policy. Again, in the absence of executive power, they provide a pattern for a variety of actors to follow over a number of years. The LHC has set itself four priority areas of concern: ethnic minority health; transport; regeneration; and inequalities. It also has an almost-explicit 5 The concept is taken from Geoffrey Dudley and Jeremy Richardson, Why Does Policy Change?, Routledge, London, 2000, p.232 6 Ibid., p.20 14 aim of opening up health policy to a wider range of influences, particularly those of policies which are linked to it. In the words of one respondent, “the black box has got to be blown open”. Consciousness needed to be changed: this is why the use of Health Impact Assessments has been to the fore. But its role, beyond being simply a ‘space’ for meeting, networking and exchanging information is unclear. It has been stymied on occasion when attempting to make recommendations to the Mayor, by mayoral advisers who believe that the recommendations would not be accepted. The Mayor’s involvement, though sometimes praised, was patchy. There was no strong desire for the Mayor to be granted extra powers over the health services, perhaps because during his campaign he had stated that he felt the Mayor should control the NHS in London. This was guaranteed to inflame the long-standing paranoia in the NHS about interference from government. It was not felt that the Mayor would be able to bring about any dramatic improvement without either “raising taxes or lowering expectations” (interview data). The LHC works primarily on the London level, reflecting its somewhat technocratic origins; the great failing of the LHC has been its difficulties linking the London, policy, level of public health work to the activities on the ground of Health Authorities, local governments, and community groups who often have the most visible health outcomes. The LHC’s natural tendency to accept subsidiarity and focus on that which can only be done at a London level, combined with the great distrust of many boroughs for anything Greater London, means that its connection with much London health work is poor—some interviewees active in local public health work could not identify any impact. Others highlighted this failing as one of time and resources. The LHC’s manifest function of assessing plans will not go on forever, as the point of the plans is that there will be a finite number of them and then implementation will start. Its latent function—of bringing together different groups and individuals to start to understand health and ways to promote health—will also change as a London policy community emerges in the new institutional context. Right now, interviewees speak of it as a “space” in which people can meet, networks form, and ideas germinate. This fact that its functions will change leads to the general consensus that the LHC will eventually exhaust its role and will have to transform itself or give way to a different kind of organisation. An advocacy coalition, after all, exists to create and institutionalise certain policies. Summary: the actors The promotion of public health in London depends overly on the behaviour of individual actors within organisations. Despite the existence of the London Health Commission and its associated networks, most of the relationships vital to the functioning of public health policy are personal. In the context of the routine reorganisations which are the hallmark of British bureaucracy, not to mention changes of political control in individual boroughs, this makes 15 the existence of effective networks through which to carry out policy extremely tenuous. Some good relationships exist, but respondents are very aware that these may not last due to circumstances outside their control: naturally, this limits the energy that actors are willing to give to building effective working. Personal relationships predominate in the absence of substantial budgets or powers for public health. The corollary of this state of affairs is that literally the only way open to public health professionals to secure effective working is to act as travelling salespeople. A project might be trialled in one borough, for instance, and a presentation made to two others, which leads to them adopting it. This virus-like working is really the only means to put into effect strategies which are written by one tier of government for other tiers to implement. One respondent caught this by stating that all the organisations involved in public health, but particularly those working at the local level, “need to be made more permeable”. London’s problem is not primarily its big teaching hospitals, the sheer number of its hospitals, its extremes of wealth and poverty, its unique social challenges, its fragmented governance or its history of distrust between layers of government. What emerged from the study is that London suffers from being under the nose of the central state. The presence of the Government, Whitehall, and the national media in the city means that London health services are under constant scrutiny and pressure to deliver targets for the short rather than long term. When the government wants quick wins, it will often sacrifice long-run public health work; outside London, entrepreneurs in public health have more leeway to preserve their programs but inside London they are easy to monitor. The trade-off between short run and long run is particularly acute in London, and all the added pressure is to focus on the short run. Thus, the greatest need for London is to have a government that can shield them from the demands of the national political scene, and set priorities for the city and its residents. The slimline, strategic GLA falls far short in this respect. A strategic body depends on voluntary cooperation. Health Authority heads who worry about “a visit to Richmond House [the DoH]” and central state servants looking nervously at Whitehall lack the latitude to cooperate as they might like. The East Midlands will show the virtues of being away from the centre; London shows the problems of being close to power. Both show the need to protect regional thinking and working from the nationalised, media-driven politics of Whitehall and Westminster. And the inability of the GLA to thus protect them suggests that strategic bodies have serious limitations when there erstwhile collaborators are busy serving somebody else. East Midlands The East Midlands was used as a comparator region for this study for a variety of good reasons. It differs from London in almost every respect, and also typifies the lack of regional 16 identity in the regions of England. It has no history of joint working; several boundaries which might easily be disputed (but have not been), and it has no one clear centre of power. There are five counties and four unitary authorities. Lincolnshire is highly rural and remote, with many associated problems; Northamptonshire, in the south, looks as much to London as to the Midlands; the three major cities, Leicester, Derby and Nottingham (each of about 250,000 people) are unitary authorities surrounded by shire counties, with the small unitary county of Rutland completing the picture. The region contains former mining areas and the tourism-oriented south part of the Peak District. The total population is some 4.2 million. The lack of regional identity or joint working, and the consequent lack of past disputes, may be a factor in the East Midlands emerging as the most effective of the English regions in terms of joint working. It certainly illustrates that regional identity is not a necessary condition of effective regional joint working. Another factor, however, was undoubtedly good management. The East Midlands Regional Assembly made the construction of an Integrated Regional Strategy (IRS) a priority early on in its life, with the result that much less effort has been spent dovetailing different plans and working at cross-purposes since then. The IRS had input from, and consequently enjoys ownership by, all the important regional stakeholders. Regional structures The interest of the East Midlands is in its advanced degree of regional working and innovation. In each case, the good luck of the region appears to be in the degree to which its institutions are “fit-for-purpose:” they are mostly fairly new, and all are in the hands of entrepreneurial leaders who see how public health policy can benefit their institutions, the regional level on which they have staked their efforts, and their particular institutions. The East Midlands contains the same regional structures as the other seven non-London regions of England. There is a Government Office for the Region, originally set up in 1994 with civil servants from four departments relocated to the region. In the last two years these four departments have been joined by representatives from Culture Media & Sport, the Home Office, and the former MAFF. GOEM has a unified directorate but no ability to vire between projects; its 2001-2002 budget of £391.61 million is confined within strict limits. Like the other Government Offices, it is an experiment to see how much benefit can come from putting civil servants together in an environment that limits departmental concerns. In the case of GOEM, it took the mission to heart, including extensive participation in regional working teams and a secondee from the NHS. emda The Regional Development Agency, emda (East Midlands Development Agency), was set up in 1999 from an agglomeration of previous national development bodies. Its budget is £91.23m; from April 2002 the budget will be subject to full virement across project heads. It 17 was tasked to produce a Regional Economic Development Strategy, which was also required to take account of environmental and social factors—a circle that proved initially difficult to square for most of the RDAs. Emda has a board of 14 people, including 3 local councillors (two of whom also sit on the board of the regional assembly). Each member is chosen, according to the act, because they have “experience of, and has shown capacity in, some matter relevant to the functions of the agency.”7 Emda also realised that public health was a way to produce significant policy outcomes, direct attention from conflictual issues onto more consensual ones, and widen the scope of its activities and its constituencies. Thus, unlike some other RDAs which adhered to a strictly business focus, emda opted to build an inclusive economic inclusion team that among other tasks built health links. emda’s participation in the IRS, its openness to social inclusion and health, and its efforts to expand relations with social partners all testify both to the activities of entrepreneurs in the region and to the value of public health as a way to bring together those who would otherwise suspect each other. The Regional Chamber The RDA Act 1998 also permitted Regional Chambers to be set up in each region of England. These were to consist mainly of local authority councillors, but at least 30% were to be drawn from wider regional stakeholders or ‘social and economic partners’. When this condition was satisfied the Secretary of State ‘designated’ the East Midlands Chamber, in May 1999, and soon after it changed its name to the East Midlands Regional Assembly. The Assembly contains at least one representative from each local authority in the region, and reflects the political balance of the region. In organisational terms it remains a part of the East Midlands Regional Local Government Association, with which it shares a budget and secretariat. The Assembly has 105 members (70 local authority, 35 partner), and meets quarterly. The ‘designation’ process entitles the Chamber to be consulted by the RDA on its economic development strategy, hence allowing regional input. But, as elsewhere, the Chamber has sought to expand its role and influence regional policy-making generally. Hence it has established relations with, amongst others, NHS Trent Region. This NHS region is not quite co-terminous with the standard East Midlands region, as it includes South Yorkshire and excluding Northamptonshire. While influence, let alone control, over the delivery of health services is not on the regional agenda, being part of the broader regional alliance of stakeholders allows the NHS—and conversely the Regional Assembly—to gain influence over adjoining areas of policy such as transport, social inclusion, and economic development. The Assembly is a classic case of an entrepreneurial body that has seen how to combine building the region and building public health in its own activities. Despite what seems like 7 Regional Development Agencies Act 1998, s 2 (2) 18 poor raw material—a region with a decidedly nebulous identity—EMRA has made a name for itself as one of the most advanced regional Chambers. Its main tool to integrate the region and demonstrate the benefits of regional working has been the Integrated Regional Strategy, or the IRS. The IRS tries to bind together as many regional stakeholders as possible in order to set a common agenda that will avoid conflicts. As part of this, NHS and HDA public health specialists were early integrated into the consultations leading up to the document and other regional partners have seen the points in common. The NHS The NHS in Trent region has been a major force for the development of public health working. The specific origins of public health work lay in the entrepreneurial activities of several major figures in Trent region, who organised a group to promote wider government involvement that incorporated representatives of other bodies. Since then, Trent and the Health Authorities have contributed by shielding public health activists and a few resources from central dictates, supporting secondments into other agencies, and trying to educate within the NHS and the rest of regional bodies. Meanwhile, the Health Development Agency, a body within the NHS, has contributed to the agenda by funding the all-important staff time that is necessary to give ideas momentum. It is difficult to separate out the HDA’s activity from the NHS proper. Resources for staff and campaigns come from both and there are not enough people for them to spat, but the dedicated staffers and small resources of the HDA have contributed, and they have the great advantage of being more remote from central micromanagement than the frontline NHS services. The current arrangements in the English regions limit the regional actors to achieving this kind of organisational permeability: without this, they can achieve nothing. Without budgets or executive authority to speak of, they can do little else. But this permeability is in fact the optimum means by which to create and deliver an effective public health policy. Public health suffers from the same marginality within the NHS that the regions suffer from within England, making the two a potentially ideal arranged marriage. The basic template is the same in London: a group of regional bodies, trying to make an impact, seek innovative policy areas in which they can make something happen while building the region as a policy space. Public health offers this, as it is a largely ignored area in which many groups can come together to squeeze added value out of their activity and sidestep old communications problems. How it works Throughout this study, we have constantly found the same pattern of regional public health development. It has two ingredients: new regional institutions with activists who want to promote the institutions and the region in new policy fields; and idea champions, spread through the NHS, who can sell public health to them as a way to build regional cooperation. 19 The actual character of the work falls into two categories: bringing together those already making policy in a region to coordinate them and squeeze out extra value—such as bringing the RDA, RA, GO and NHS together in the East Midlands—and using the regional level to build the coherence of and lend assistance to the local level where local government, voluntary organisations, social enterprise, Health Authorities and others produce on the ground results. This formula requires some basic institutional prerequisites, which this section lists. A successful public health policy requires all of them; we argue that transferring extensive public health responsibilities to elected regional assemblies is the best way to make public health in the regions a success because it will best mobilise the four mechanisms of shields, soft money, secondees and networks. Soft money Public health can be most effectively addressed through ‘soft’ or unhypothecated money. Under the current government this kind of money is in very short supply. Most of the money used by the regional bodies towards public health is in fact soft money, but the lack of it is an inhibitor. One respondent indicated that a great benefit of more money would simply be the ability to complete programmes more quickly. It is very difficult for public health to find money which is not either tied to specific programmes or to achieving performance indicators. Together with the proliferation of health action zones, plus other action zones, the disjointed-ness of which was summarised by Reaching Out,8 the result is that local and regional organisations lack the opportunity to innovate locally due to being tied to central management targets. The London NHS regional executive needs to have more freedom to innovate, according to one respondent. This lack of freedom also hinders regional joint working: regional NHS offices cannot undertake programmes with other organisations due to lack of financial or managerial leeway. As such, when soft money does become available, it often has an apparently disproportionate effect. In East Midlands and NHS Trust, soft money becomes available through a variety of grant mechanisms. The Health Action Zones, the Single Regeneration Budget and European Union grants both encourage joint working and result, when successful, in the availability of funds which are not tied to central targets and hence central control. The disadvantages of this form of funding is that it has a limited shelf-life, and thus it can be difficult to extract commitment either from partners or from the staff hired to work on it. But its existence in the first place is a vital way of attracting the attention of busy people with innovative ideas, and creating new coalitions and policy networks on the ground. It can thus be crucial to formulating a regional policy for the people of that region, and is where regions could make their biggest and most positive mark. 8 Cabinet Office (Performance and Innovation Unit), Reaching Out, February 2000. This report analysed the degree of joining-up of governance in the English regional structures. 20 Secondments The use of secondees is a common feature of the public health agenda. It represents a creation of soft money where it was thought not to exist—that is, a secondment is a significant form of support in kind for an under-funded public health initiative. But even if substantial resources were available, evidence indicates that secondments are a useful way of sharing organisational experience and of achieving permeability. Even a part-time secondee can introduce new ideas to an organisation lacking the time on its own to give (say) public health issues serious consideration. Secondees, coming from outside an organisation, are not bound by the culture and conventions of that organisation: they are far less likely to fear for their future if they speak up over inefficiencies, and are more likely to have flexible time enabling them to make links with their erstwhile colleagues. They also may well be more likely to be creative and to ‘think outside the box’; otherwise they would not have gone on secondment in the first place. Secondments are more eagerly used in the East Midlands, reflecting their greater ability to experiment and the less intense pressure on them to deliver. In London, they are under-utilised because the high-level networks that pave the way for secondments are still weak (despite the LHC) and due to the greater pressure on London organisations. It is vital that any reform of the regional public health agenda should take into account the value of secondees in moving agendas forward and dedicating time and effort to the issue, and build in funds for secondments between organisations. Regional competence must not consist simply of moving staff around within familiar job titles. Networking The London Health Commission is a formalised network. It links some forty organisations through meetings of their chief executives, thus encouraging cross-fertilisation of ideas. Theoretically those who attend will take ideas back to their own organisations. But, according to two respondents, there is scant evidence of this happening—perhaps not surprising given the pressures on chief executives. Partnership fatigue can lead to chief executives giving up on joint working if no tangible benefits are evident, and particularly if no money is available to ease the process. Secondees are a more promising means to this end. In the absence of clear executive power over public health, the network(s) created around the London Health Commission are a useful and efficient means of policy integration—in theory. Chief executives from many different organisations can meet, share problems and information, and find out about initiatives which dovetail with their own. But the evidence that this is leading to alteration of policy by any of these organisations is thin. Networks in London, as in the rest of the country, will be threatened by the upcoming reorganisation of the NHS into 4 super-regions, 28 strategic health authorities and nine directors of Public Health relocated in Government Offices. But they are likely to be less 21 threatened in London than elsewhere. The London regional office will be the only region to survive unscathed. Underneath it will be five strategic health authorities, in place of the current 16 health authorities. Staff faces in the new system will often be familiar from the old one: relationships will be disrupted, but not fatally. None of this, however, should disguise the almost fatalistic tedium of yet more reorganisation expressed by the respondents. A further advantage for London is that most Primary Care Trusts are, or will soon be, coterminous with borough boundaries. This makes joint working at borough level (in principle) much easier. The sheer size and legacy of disjointed working in London has meant that the construction of top-level networks is slow. In the East Midlands, top-level executive networks have been able to function and build trust and experience. There is far more to gain in that region, with so little benefits derived up till now from regional working, and perhaps crucially, the relevant personnel have seen this. Whether or not the elimination of Trent NHS authority in the upcoming reorganisation will reverse this progress remains to be seen, but the omens are not good. Shields The use of ‘shields’—governmental organisations defending programmes from interference from the centre—was noted in the East Midlands by Greer in The Real Regional Health Agenda. The use of shields was much less noticeable in London. In part this derived from poor relationships between the relevant actors. There was less inclination by the NHS, for instance, to shield new developments proposed by any of their partners, and conversely (a chicken and egg situation) little will to propose new ideas. The King’s Fund, a well-resourced and independent organisation, is able to fill this gap to a considerable extent: it has driven the development of the London Health Observatory by giving it both kudos and office space. But no form of London governance can avoid London’s extreme visibility. It is the UK’s capital, and seat of a highly centralised government. Many opinion formers live there and are directly affected by the quality of services, or aware of local decisions which would sink without trace in most parts of England. The greater impact on Whitehall of issues specifically of London governance encourage greater Whitehall interference generally. London Underground is the most potent example, but in another instance, the Mayor was offered no role in protection measures for London following the terrorist attacks of 11 September 2001 in New York. London is also the hub of an extremely centralised national media. Mistakes by health authorities in the other English regions will be picked up by local or regional press; mistakes by London health authorities are often picked up by the national press. It is easier for London-based media to report on events on their doorstep. Intense media and government interest, and a tendency to government interference, makes life uncertain for many 22 organisations. There is very little possibility of shielding for London as a result, and, perhaps in turn, few of the large organisations attempt it. In the East Midlands, distance from London, the metropolitan preoccupations of media and politicians, and the leadership of a few powerful individuals in the NHS and government have been able to shield regional health work, thereby making possible the achievements to date. Shielding is the crucial reason for giving public health to the regions. Public health involves integrating health concerns into a host of different activities. If it were a regional activity, it would be a centre of a broad integrated agenda. Without a regional shield to defend public health, as one interviewee put it, “it gets kicked off every time by waiting lists”. Safely within the regional body, public health can be disconnected from statistics on waiting lists. Why Devolve Public Health? Devolution to the regions of England is a problematic subject for this government. A proliferation of performance indicators, close interest in local government affairs and tight discipline have been hallmarks of the Blair governments. In particular, the second Blair government stressed its commitment to improvement of public services: it would be very surprising if the regional White Paper advocated the decentralisation of those services and the consequent loss of control over them by central government. On the other hand, the same government has brought devolution to Scotland and Wales, and has long promised unspecified forms of elected government for England’s regions. And it has always specified that elected regional government would need to be approved by referendum. Thus the White Paper will necessarily tread a thin line between producing a model of regional government strong enough to attract support in a referendum, and a model which is not so strong as to strip central government of responsibility for public services. Therefore, it will have to draw up a list of powers which could reasonably be devolved to elected assemblies. Public health is a leading candidate. In the Introduction, we mentioned four reasons why public health would function better as part of a range of powers enjoyed by regional assemblies. The rest of this section details why public health is an excellent candidate for devolution. 1. Regional public health would build on existing success Devolving public health to the regions would build on the existing regional successes studied in this report. Regional public health, almost entirely on the backs of activists and idea champions in and outside of the NHS, has begun to show local results and, perhaps more importantly, change the way policymakers think about public health and the regions. As we have argued above, much of this is because of the development of strong regional policy networks and webs of secondments with slack resources and shielding from central 23 pressure. An elected regional assembly could enhance all of the above. Furthermore, it would have incentives to throw its weight behind public health, given the opportunity public health offers to bring together different regional players and the prominence it would have among the likely powers of elected assemblies. This would all be further reinforced by the regions’ competencies, as public health would for them be a major responsibility, and it would be disconnected from the constant pressure on waiting lists that damages NHS public health working. There is one fundamental reason that public health is such a good candidate to move into the regions. As we have said, public health is an inherently interstitial, entrepreneurial, viscous policy field. As such, it works to overcome departmental and functional divisions, creating neutral spaces for new thinking and eroding old barriers. It brings policymakers from different fields together to discuss topics that they had not discussed before—and therefore can give them a common cause in many cases where they previously had none. It thus depends on smaller policy communities and more trust than national policies. It also depends on local conditions that allow actors to forge partnerships that would seem far to specialised to national policymakers: programmes we studied such as a bus improvement programme in rural Lincolnshire or integration of mental health services and disability benefits for a troubled population in London can be identified and can create new networks far more easily in a region than in the central policy departments. Thus, for its adaptability, small size, and high trust, the region is an effective level on which to promote public health. 2. Regional public health would fit with a broader role for regions Most of the conversation about what regions might or might not do is about the way they can give shape, coherence, and democratic accountability to existing policies. Nobody proposes the sort of power over policies that the devolved assemblies (or federal states of other countries) have; the assumption is that the rights and responsibilities of every person in England must remain equal. Instead, the debates about regionalism are mostly about how to give people in the region a say in the way policies affect them, about how to make policies work better by integrating them and adapting them to each region, and about how to create intelligent regional strategies that can focus and guide politics and policy. Thus, debates about regionalism stress the possible roles of regions in planning, in social inclusion, in environmental and transport planning decisions, and in economic development. All of these are fields with strong practical similarities to public health. On one hand, they all involve making decisions and adapting policies to strategic goals, rather than direct service provision. On the other hand, they all could contribute more to people’s welfare if they were to take health into account (and some, such as land use planning, economic development, and social inclusion, would work better if everybody realised just how big and economically important the NHS is). Public health could be part of regions that build their expertise and 24 focus on shaping policy regimes to suit their voters and that add great value by integrating concerns and goals across policy domains. Also, it is likely that elected regional assemblies will have full, or at least some, freedom of virement over their budgets. Without that freedom, regional assemblies’ reach will be extremely limited. With that freedom will come the advantages of ‘soft money’ set out above—along with the potential to integrate public health initiatives with transport, environmental, and cultural policy, amongst others. A regional government would be able to move funds about between these areas and/or to construct joined-up policy as it saw fit. If the Public Health Observatories’ budgets were to be subsumed within the Assembly, the capacity to join up would increase still further. 3. Regional public health would integrate health with other policies The divorce between health and health services, common to most countries, is particularly great in Britain due to the particular history and structure of its health services. The NHS is a health services organisation, with public health staff whose roles vary from epidemiology and disease screening to the networking and policy described here, while many of the policies with the most influence on public health are spread across organisations as diverse as local authorities, the voluntary sector, quangos and Next Steps agencies, and Whitehall departments. Even finding the right people to form a policy coalition takes some work in this environment. Then comes the mistrust and misunderstanding that can stymie work, which can only be expected when one thinks about the histories of the different actors. It is not surprising that health and public health in Britain are separated; it is instead surprising that the activists we interviewed could create the networks and policies. Regional public health could surmount this historic gap by bringing together many of the actors under one umbrella, and giving a stable contact point for the rest. Planning, environment, social inclusion and other functions that might go to the regions could work together closely and with the same democratic accountability, while others such as the Primary Care Trusts or Next Steps agencies could know who to talk do and could have an interlocutor with real weight, resources, and stability. The presence of public health as a specified regional competence would raise its profile in the eyes of the electorate and central government, and hence regional partners. Regional executives would be able to give stronger leadership in the field of public health, transforming its Cinderella status. They would have incentive to do so, as it might be one of their most important and high-status powers. And the presence of more resources and greater political commitment would enable initiatives to be completed much more quickly, to greater benefit. 25 4. Regional public health would fit with the developing structure of public health in the NHS With the upcoming abolition of the NHS executive, Regional Directors of Public Health will be transferred to the Government Offices for the Regions, cutting across the new regional structures proposed for the NHS. The rest of the reorganisation pushes powers down to the local level with Primary Care Trusts doing most work and commissioning, and up to the level of three giant new Health and Social Care regions (plus London) and the centre, which wants a “clear line of sight” to the frontlines. Public Health will in this case be the odd one out, the only part of the NHS on the regional level. This reorganisation means that many of the criticisms of regional public health have already been met. Critics can charge that by dividing public health into regional policy teams and the gritty work of public health specialists in the PCTs, the field will be split and distorted. If these charges are likely be proved right, then it is already too late. On the other hand, the integrative function of public health could be carried out even better as a major, professional, high-status part of a regional assembly’s work rather than as another, small, component of the work of the Government Offices. Also, one major worry is that small, pressured, and GP-dominated PCTs will forget public health, and the Department of Health and Secretary of State will let them. Having a strong regional lobby to support public health funding and to work with PCTs—and regional politicians who would perforce have to make a mark in public health—could easily be a boon to public health in the local areas that outweighs any disadvantages of splitting the profession. Blueprints The goals of any regionalisation of public health should be: To facilitate regional joint working by bringing decisions down to a more flexible and accountable level of government; To raise the profile of public health and build on its success by making it a key part of the assemblies’ powers; To make the most of—and build up—the assemblies by giving them a field that combines real issues (cancer, heart disease, teen pregnancy) with ample need for strategic thinking; To use the fact that both public health and regions work best when they can bring resources to bear on promoting, through networks, policy integration and new ideas that more effectively use existing resources and institutions. The basic outline for a regionalisation of public health competencies is relatively simple, given the changes in regional and health system design in the last year. We propose: 26 That the integrated public health teams currently being formed for transfer to the Government Offices be transferred to the elected regional assemblies upon their constitution; That the budgets of these teams be transferred as well as the available funds for regional public health promotion within the Department of Health and NHS Executive; That the regional assemblies have funds that can be vired between purposes and used as seen money or grants, and that they have a responsibility to use them to promote public health as well as their other goals; That the regional assemblies have additional soft money funds available as part of their start-up grant, in order to more effectively shape their regions’ policies; That central departments may not introduce specific public health grants or programs in a region without the regions’ “signing off” on the plans, and that concordats similar to those for the existing devolved bodies regulate their relations. The public health competency There must be a core regional public health competency that encompasses, at a minimum, responsibility to promote the long-term health of the population through policy coordination and targeted programmes while in the short run working to reduce inequalities, improve morbidity rates, and incorporate environmental health. The government to date has usually just given lists of areas that are key public health foci, since as coronary heart disease and inequalities of access. We suggest that a definition of public health err toward the broad for regions, and bear down on policies related to improving disease rates and quality of life; reducing need for acute hospital services and long-term or social care through disease reduction; promoting education in the population and among institutions; and supporting policies, policy integration, education and projects that will achieve these goals. The definition should also make clear that some public health competencies will remain within the NHS, in the hands of the PCTs; thus, the NHS and Department of Health should be equals or leaders in improving access to medical treatment; emergency public health measures; all forms of disease screening; care in the community; and ground-level disease prevention work. The regions will be able to work with the PCTs because of their broader view, larger networks, and above all larger and more fungible resources. Transfer of staff and budgets The new regional public health competencies should come with the transfer of staff from the Integrated Public Health teams now being created in the Government Offices. Given that these teams will have been created by the time any assemblies start work (in fact, they are already planning their moves into GOs), their borders and budgets will already be set by the time any regional devolution happens. These experienced public health officers should be transferred with their conditions of pay and work in the NHS as a minimum, and will form the core of expertise and political skill around which elected regional representatives can 27 build their activities. The regional public health observatories are likewise a growing core of skills and knowledge. They and their staff and budgets should likewise be devolved to the regions where they can have the most impact. More importantly, the government should devolve granting authority for public health to the regions: funds from HAZs, grants associated with HIMPs, social inclusion grants with an important health component, including New Deal for Communities and Single Regeneration Budget funds should be transferred as a block grant to the regions. This budget, combined with other budgets transferred to regions, should probably be allocated according to a basic formula that takes into account need. Otherwise, the funds will be too politicised. While the regions should receive such funds, because they are necessary to do the regions’ work, additional funds from the government could work via regions and the centre negotiating a formula and an expected outcome, and the centre can thus see if the additional funds invested in the partnership are working. Above all, regions should have the power to vire funds. For example, a regional assembly might have five major competencies: social inclusion; public health; transport planning; economic development and neighbourhood regeneration. All five of these are closely interdependent (a good bus system dictates whether the poor can apply for distant jobs, unemployment strongly influences substance abuse, substance abuse ruins neighbourhoods and damages health indicators, etc.). A regional assembly, unlike today’s interdepartmental initiatives, could design programs using funds to create integrated solutions that reflect the region and its people’s preferences. Regions should coordinate with the centre and have a veto over central public health initiatives We do not propose many formal obligations for regional competencies. This is because the point of regional devolution is to allow regions to vary in accordance with their needs, approaches, and democratic preferences. As the unhappy recent history of local government demonstrates, putting too many technocratic constraints on an elected body defeats the purposes of having it, namely innovation, adaptation, and democratic accountability. One of the best arguments for having an elected regional government is precisely that it can shield public health from central micromanagement. Thus, the only substantive recommendations we make are that the regions should be asked to sign concordats with the Whitehall departments involved in the relevant policy areas, namely social exclusion, long-term care and mental health, environment and planning, and education. These concordats would above all serve to prevent gaps and conflicts between regional and national public health priorities. 28 Instead, the important matter is that the design of the regional assemblies create something akin to the equality that the present devolved bodies have when facing Whitehall. The regional governments should have veto power over central public health spending in their regions; otherwise nobody need listen to them and they will not be able to fulfil their democratic or technical goals. Whitehall should not be able to run parallel programs in regional public health competencies; the experience of devolution in Northern Ireland, Scotland and Wales demonstrates that these kinds of clashes can be avoided or resolved, and that devolved and central governments can negotiate the exact divisions of powers. What is necessary to establish a basis for an amicable division of labour, however, is an obligation for Whitehall to listen to the devolved governments. Given that the NHS will retain extensive public health competencies in the PCTs and in the national level, the worry is that regions will be squeezed out. That is why we propose that the regions have, effectively, a veto power over public health projects directed from the centre. It is unlikely that regional governments would object to greater resources in their public health systems, but they must have some way to prevent their being sidelined by the centre. Furthermore, policy will work better if there is integration, and since it would be a regional core competency and optional for Whitehall, it is the region that must take pride of place. As a route to making the most of policies in general, and to specify what will need to be in the concordats, regions could produce public health plans in line with their other competencies, but this (while very likely) need not be obligatory. They might also develop closer links with the new strategic health authorities to ensure co-ordination of activity, possibly including a more general health strategy for the region which the NHS would sign up to. Ideas for the regions The core of regional public health will be in its ability to shield and promote entrepreneurs by creating a core around which networks form, and funding worthwhile policies. However, there are other opportunities in the regional devolution of public health. There should be a distinct scrutiny / policy development committee for public health. This would be another opportunity for innovation. The appointments for this committee should recognise the cross-cutting nature of public health issues, and ensure that the members form a broad representation of the membership of other relevant committees—transport, environment, culture, housing.9 This will maximise the flow of knowledge and different perspectives on public health. 9 That is to say, the members of the Public Health committee would all be expected to be members of at least one other committee. This argument assumes that regional assembly ‘back-benchers’ would need to hold membership of at least two committees, due to regional assemblies’ numbers being small. 29 It is also likely that regional assemblies would be able to make innovative use, if they so chose, of special advisors, possibly of co-opting members of scrutiny / policy committees. There might be a useful role here for a body resembling the London Health Commission. That body has been somewhat frustrated in its work, according to some respondents, because of its large size (40+) making it unwieldy. It is also not clear whether its role is to advise on existing policy or to adopt a more proactive role. The latter could only be a beneficial means to bringing expertise on board. A slimmed-down version of the London Health Commission, with perhaps a dozen members, and with Assembly secretarial support, could provide a more focused source of advice. It would also be a vital means by which NHS senior management would maintain relationships with the regional assembly: it would be painfully easy for the two to arrive at loggerheads very early on. A clear way of reintroducing links between the region and the sub-region—and at the same time increasing the democratic accountability of the NHS—would be to oblige some of the board members of each SHA to be drawn from the members of both the relevant regional assembly and the relevant local authorities. In the case of the regional assembly, the members could also be drawn from the public health policy / scrutiny committee (though, together with the proposal above, this might lead to overload): and in the case of local government, from any health scrutiny committee that had been set up by local authorities. An early useful task for the Public Health directorate would be establishing relations with any existing scrutiny committees, and feeding their findings into an archive available for the regional assembly and local authority to use as a database of past knowledge. Unelected chambers In those regions which do not vote, in the medium term, for elected assemblies, something approaching the above proposals could be constructed alongside, or perhaps within, the Regional Chamber. Mark Sandford, in Further Steps for Regional Chambers 10 , proposes relocation of the Public Health Observatories within the ambit of the Regional Chambers, including the ability to vire their funding. A distinct Public Health body could be formed, similar to the proposal above for a more focused Health Commission, with a shared secretariat, a minimal (probably mostly seconded) staff, to co-ordinate public health partnerships and produce a health strategy for the region. That paper also suggested a role for Regional Chambers in the proposals for joint local authority scrutiny committees to hold health authorities to account: it is likely that a public health body of this type would have a useful role to play in regional or sub-regional scrutiny. Most likely Directors of Public Health would remain within Government Offices, but these could still establish effective relationships with Chambers and Strategic Health Authorities. 10 Mark Sandford, Further Steps for Regional Chambers, Constitution Unit, December 2001 30 The proposals for elected members to sit on SHAs might still be implemented, though probably fewer members would be able to do so. In regions which have not voted for elected assemblies there would still be a role for public health. We suggest that the existing Regional Assemblies be charged with a public health planning and scrutiny function. The objectives of this activity would be: to ensure that the development of regional political networks includes health, both as public health and the NHS services proper, and on a voluntary basis from the start rather than as a later imposition on the health service; to ensure that public health takes an appropriate place in all regional activities by both writing it into regional plans and creating awareness of how and why health concerns matter; to take full advantage of the transfer of public health into Government Offices by both solidifying older links into the NHS and creating new ones in the Government Offices; to thereby build the networks and consensus necessary to incorporate full regional public health powers into any later elected regional assembly. The specific obligations would be: The regional assemblies, in consultation with all of their partners, but specifically the Government Offices, RDAs, regional NHS services on all levels, shall formulate public health plans for their regions; These plans shall focus not on health services, but on public health and the wider determinants of health. Existing NHS services—especially PCTs—will be expected to contribute primarily through their knowledge of the problems that their populations face and their suggestions for how broader government integration could help solve them; The plans will, subject to government approval, supply the broad work program for public health officers as they move into the Government Offices; The plans will provide the basis for integrating the strategies of RDAs into public health. Resource requirements would be (in each region): One DoH civil servant (or NHS, or HDA) willing to take the secondment to the RA; Expenses for the secondment of up to four more staffers between the GO, RDA, RA, local government association and NHS on a limited part-time basis over two years; Expenses necessary for the program of regional consultations organised by the secondee; A contribution to the administration of the office in which the secondee works, and a personal assistant to maximise the secondee’s efficiency. 31 Conclusion The regional government agenda of the Labour Party, perhaps contrary to expectations, is uncovering plenty of hidden energy in regional institutions. Different government and non-government organisations are finding real benefits in the greater freedom to work together—even though that freedom is still severely limited and the staff time and budgets available are still extremely small. While in London, visibility has hindered the process to some extent, as a capital city it is only to be expected that central government will find it difficult to avoid meddling in regional or local affairs. Our proposals in the final section of this paper are made to suggest how the emerging agenda of elected regional government can best harness this energy. One of the continuing concerns of government, at least on paper, has been to overcome the ‘silo mentality’ of Whitehall, where inter-departmental co-operation is eccentric if not unheard of. Greater devolution of regional competence enables this obstacle to be bypassed, and, if carried out in harmony with processes which have already worked well, will prevent regional government from repeating national government’s shortcomings. Bibliography Department of Health, Shifting the Balance within the NHS: Securing Delivery, London, 2001 Geoffrey Dudley and Jeremy Richardson, Why Does Policy Change?, Routledge, London, 2000, p.232 EMRA / EMDA, England’s East Midlands: A place where Life Works, Melton Mowbray, 2001 Scott Greer, The Real Regional Health Agenda: Networks, Soft Money, and the Wider Determinants of Health in the East Midlands, Constitution Unit, 2001 Brendan O’Leary, “London Governance: Past,. Present and Future”, in eds Michael Hebbert and Tony Travers, The London Government Handbook, Cassell, London, 1988 Mark Sandford, Further Steps for Regional Chambers, Constitution Unit, 2001 32 Related publications by the Constitution Unit The Constitution Unit’s publications have proved and invaluable source of information for policy makers, practitioners, academics and students. All our work has a sharply practical focus, is concise and clearly written, timely and relevant to policy makers and practitioners. Further publications linked our regional government and devolution and health programme are listed below. Devolution and Health: Final Report, June 1998 Available from The Constitution Unit: 020 7679 4977 £10 Devolution and Health: First Annual Report, February 2000 £10 Available from The Constitution Unit or free on line from The Constitution Unit’s website: www.ucl.ac.uk/constitution-unit/d&h Devolution and Health: Second Annual Report, June 2001 Free on line Available from The Constitution Unit’s website: www.ucl.ac.uk/constitution-unit/d&h Unexplored Territory: Elected Regional Assemblies in England, July 2001 Available from The Constitution Unit: 020 7679 4977 £10 The Real Regional Health Agenda: Networks, Soft Money, and the Wider Determinants of Health in the East Midlands Free on line Available from The Constitution Unit’s website: www.ucl.ac.uk/constitution-unit/d&h Divergence and Devolution, November 2001 Available from The Nuffield Trust: www.nuffieldtrust.org.uk £5 Further Steps for Regional Chambers, January 2002 Free on line Available from The Constitution Unit’s website: www.ucl.ac.uk/constitution-unit Devolution and Health Quarterly Monitoring Reports, published each year in February, May, August and November Free on line Available from the Devolution and Health pages of The Constitution Unit’s website: www.ucl.ac.uk/constitution-unit/d&h The Unit publishes a quarterly newsletter, The Monitor, together with a regularly updated list of publications. To go on the mailing list, please contact the Unit at the address above, or visit our website at: http://www.ucl.ac.uk/constitution-unit. For further information or to order publications, please contact the Unit on: Tel: 020 7679 4977 Fax: 020 7679 4978 Email: constitution@ucl.ac.uk An order form and full list of publications can be found on our website at: http://www.ucl.ac.uk/constitution-unit